Download Modern neuroscience is based on ideas derived

DEAD TISSUE, LIVING IDEAS: FACTS AND THEORY FROM NEUROANATOMY
Helen Barbas
Modern neuroscience is based on ideas derived
from neuroanatomy dating back to the neuron
doctrine in 1888, when Ramon y Cajal
demonstrated with the Golgi stain that the nervous
system is not a syncytium but consists of
individual neurons (DeFelipe and Jones, 1988).
When the electron microscope was introduced
about 50 years later, the gaps between neurons
could be seen at high resolution and neuroscience
acquired a powerful tool to probe the intricate
structure of the synapse, and more recently its
modification in learning and in disease.
The organization of the cortex into layers and
columns is fundamental for neural function.
Brodmann relied on differences in the shape and
arrangement of neurons in cortical layers to divide
the cortex into architectonic areas (Brodmann,
1909). Superimposed on the laminar (horizontal)
organization, is a vertical organization into columns
of neurons with similar features (Mountcastle et al.,
1955). This dual organization of the cortex
considerably increases its computational capacity
(Grossberg, 1999), providing the structural basis
for segregation and interaction of inputs and
outputs, local inhibitory control, and parallel and
distributed processing.
Columns mapped in physiologic studies were
first thought to span the entire depth of the cortex.
Evidence that this is not true in all cases emerged
when a simple histochemical procedure for
cytochrome oxidase fractionated columns of the
primate primary visual area by intensely labeling
‘blobs’ in layers 2-3, setting them apart from sites
with low intensity. A re-evaluation of the response
properties indicated that neurons in blobs had
distinct physiology as well (Livingstone and Hubel,
1984).
It wasn’t until the introduction of neural tracers
that columns could be seen in brilliant clarity and
their organization understood, as ocular dominance
columns in the primate primary visual cortex, or
barrel fields in the rodent somatosensory cortex.
Neural tracers replaced the pioneering but difficult
and limited ablation-degeneration mapping
methods, and offered exciting new possibilities. No
other technique has comparable power and
flexibility to show at once the spectrum of inputs
and outputs of small or large brain areas, a column,
layer, or single neurons. Using tracers we learned,
for example, that connections between any two
structures are generally reciprocal. Initially all but
Cortex, (2004) 40, 000-000
ignored in functional studies, it is now clear that
reciprocal connections have a fundamental role in
all neural systems, ranging from simple sensory
perception to complex cognitive processes [for
review (Barbas et al., 2002)].
Neural tracers have made it possible to study
the interactions of prefrontal association areas,
which were not easily amenable to physiologic
study and remained ‘silent’ until recently [review
in (Goldman-Rakic, 1996). A conceptual advance
was made when anatomic data showed that the
prefrontal cortex, long thought to be the seat of
cognition, also has a limbic component, a system
classically associated with emotions (Nauta, 1979;
Yakovlev, 1948). The limbic cortex, situated on the
medial and basal surfaces, has expanded in primate
evolution along with the association cortices and
maintains strong connections with them,
inextricably linking areas associated with cognition
and emotion [for review (Barbas et al., 2002). This
evidence challenges the classic idea of Plato that
thoughts and emotions are entirely separate. In fact,
disconnection of pathways associated with
cognition and emotion likely is at the core of
psychiatric diseases characterized by inability to
attach the appropriate emotion to a situation. Such
pathology disproportionately affects limbic areas,
and by extension, feedback communication, the
predominant pattern of projections of limbic areas
(Barbas et al., 2002).
In spite of its empirical and conceptual
contributions, neuroanatomy is often underestimated
for its power to understand neural function. This
perception
stems
from
a
fundamental
misunderstanding that information obtained from
‘dead tissue’ is limited, as a colleague physiologist
once suggested. For him a change of mind came
fortuitously, when we were examining histological
slides from the brain of one his monkeys and saw
that a large part of the lateral geniculate nucleus on
one side had degenerated. Paradoxically, the
monkey’s partial blindness had escaped notice
during the behavioral training and physiologic
recording, but was now inferred from a set of slides.
Beyond such perceptions, however, problems
exist in neuroanatomy, weighed down with long
lists of terms, confusing variations in maps, and
architectonic borders that most cannot easily see.
Can the problems be resolved? I believe so. Take,
for instance, variations in maps. We now have a
host of molecular markers that are differentially
2
Helen Barbas
distributed in the cortex, making it possible to
construct quantitative profiles and characterize
architectonic areas objectively. Figure 1 (left)
shows one such example, where a dozen
quantitatively assessed architectonic features were
considered simultaneously using multidimensional
analysis. Prefrontal limbic areas clustered to the
left, and the eulaminate to the right, because they
differ structurally. Objective approaches eliminate
the need to rely on subjective analysis and can
resolve differences in maps in the literature.
Reliable maps are needed, now more than ever,
to localize activity and interpret functional imaging
studies in humans. However, the value of maps
transcends the need to localize because structure
affects the pattern of connections. Thus, systematic
variations in architecture underlie the graded
laminar pattern of corticocortical connections [see
(Barbas et al., 2002)]. Patterns of connections have
variously been interpreted to reflect the direction of
processing in sensory areas, or the distance
between connected areas [for review (Felleman and
Van Essen, 1991)]. However, just as columns are
not unique to sensory areas but represent
fundamental organizing units in the cortex (Fig. 1,
right), so is cortical structure, whose systematic
variation can be used to explain and predict the
pattern of connections equally well in sensory,
association, and limbic cortices.
A frequently voiced criticism of neuroanatomy
is that it’s descriptive. But accurate quantitative
description is the sine qua non for the life sciences,
whether it’s labeled boutons, firing properties of
neurons, or expression of genes. Darwin used it,
and so did Ramon y Cajal, and the mappers of the
human genome. Darwin’s meticulous descriptions
of finches, barnacles and rock formations were
necessary for the inductive process, giving rise to
the theory of evolution. Quantitative empirical data
provide the opportunity to reveal relationships,
derive principles, perform computations, model,
predict normal function and dysfunction in disease.
And this is where the excitement and challenge lies
in neuroanatomy – fortified with quantitative tools
and analyses it is at the core of neuroscience.
REFERENCES
BARBAS H, GHASHGHAEI H, REMPEL-CLOWER N and XIAO D.
Anatomic basis of functional spcialization in prefrontal
cortices in primates. In Grafman J. (Ed) Handbook of
Neuropsychology. Vol. 7, 2nd edition. Amsterdam: Elsevier
Science B.V., 2002, Ch. 1, pp. 1-27.
BRODMANN
K.
Vergleichende
Lokalizationslehre
der
Grosshirnrinde in ihren Prinizipien dargestelt auf Grund des
Zellenbaues. Leipzig: Barth, 1909.
DEFELIPE J, JONES EG. Cajal on the cerebral cortex. An annotated
translation of the complete writings. New York, Oxford:
Oxford Univ. Press, 1988.
FELLEMAN DJ, VAN ESSEN DC. Distributed hierarchical processing
in the primate cerebral cortex. Cerebral Cortex, 1: 1-47, 1991.
GOLDMAN-RAKIC PS. Regional and cellular fractionation of
working memory. Proceedings of the National Academy of
Sciences of the United States of America, 93: 13473-13480,
1996.
GROSSBERG S. How does the cerebral cortex work? Learning,
attention, and grouping by the laminar circuits of the visual
cortex. Spatial vision, 12: 163-185, 1999.
LIVINGSTONE MS, HUBEL DH. Anatomy and physiology of a color
system in the primate visual cortex. Journal of Neuroscience,
4: 309-356, 1984.
MOUNTCASTLE VB, BERMAN AL and DAVIES P.W. Topographic
organization and modality representation in first somatic area
of cat’s cerebral cortex by method of single unit analysis.
American Journal of Physiology, 183: 646, 1955.
NAUTA WJH. Expanding borders of the limbic system concept. In
Rasmussen T and Marino R (Eds) Functional Neurosurgery.
New York: Raven Press, 1979, pp. 7-23.
YAKOVLEV PI. Motility, behavior and the brain: Stereodynamic
organization and neurocoordinates of behavior. Journal of
Nervous and Mental Disorders, 107: 313-335, 1948.
Dept. of Health Sciences, Boston University, Boston MA., USA. [email protected] and
NERPRC, Harvard Medical School, Boston, MA
Fig. 1 – Analysis of architectonic features and pattern of connections in the prefrontal cortex in the rhesus monkey. (Left)
Quantitative architectonic features of prefrontal cortices (neuronal density, glial density, thickness for layers 1, 2-3 and 5-6, and for the
calcium binding proteins parvalbumin, calbindin and calretinin for layers 1-3) were considered simultaneously using nonmetric
multidimensional scaling (NMDS). The plot indicates the relative similarity of prefrontal areas according to normalized laminar profiles
across the quantitatively obtained experimental measures. In this analysis the limbic prefrontal areas cluster to the left and the
eulaminate to the right. (Right) Columns formed by the termination of axons (white) emanating from area 46 and terminating in area 12
of the prefrontal cortex. Scale = 1 mm. (The figure on the left was adapted from Dombrowski, Hilgetag and Barbas, Cerebral Cortex
11:975-988, 2001.)